Mathematical modeling for a wastewater treatment plant in city of Velika Gorica

Obrada otpadne vode s aktivnim muljem izrazito je složen proces, a nestalnost sastava i protoka otpadne vode te vremenski promjenjive reakcije mikroorganizama uvjetuju njegovu nelinearnost i dinamičnost. S tim u skladu i njegovo je optimalno vođenje otežano, ali pravilno izrađeni model mogao bi poslužiti kao rješenje problema. Stoga su u ovom radu izrađena dva modela za opis tog procesa na uređaju za obradu otpadnih voda grada Velika Gorica ("VG Vodoopskrba"), a s ciljem predviđanja vrijednosti KPK u izlaznoj vodi. Izrađeni su modeli: viševarijantni linearni model sa 13 varijabli i model lokalne linearne regresije sa 13 varijabli. Od ova dva modela uspješniji u predviđanju ponašanja procesa (korelacija 0,62) pokazao se model lokalne linearne regresije.Wastewater treatment with activated sludge is a very complex process, and nonstationary behavior of pollutant concentration and time variation of microbial activity results in nonlinearity and dynamic character of the process. Based on these characteristics optimal process control is difficult, and mathematical model could be applied to solve the problem. For this reason two mathematical models are developed for wastewater treatment plant in city of Velika Gorica ("VG Vodoopskrba") with the goal to predict COD in the effluent. Applied are the models with multivariate variables: linear and piecewise linear regression (PLR) models with 13 variables. The model by PLR method proved to be more accurate yielding the correlation coefficient of 0,62